The present invention generally relates to process control, and more particularly to methods of volumetrically controlling a mixing apparatus.
The following paragraphs contain some discussion which is illuminated by the innovations disclosed in this application, and any discussion of actual or proposed or possible approaches in this section does not imply that those approaches are prior art.
The following applications filed concurrently herewith are not necessarily related to the present application, but are incorporated by reference herein in their entirety:
“Methods of Determining a Volumetric Ratio of a Material to the Total Materials in a Mixing Vessel” (U.S. application Ser. No. 11/323,831, filed simultaneously with the effective filing date of the present application)
“Systems for Determining a Volumetric Ratio of a Material to the Total Materials in a Mixing Vessel” (U.S. application Ser. No. 11/323,323, filed simultaneously with the effective filing date of the present application), and
“Methods of Volumetrically Controlling a Mixing Apparatus,” (U.S. application Ser. No. 11/323,324, filed simultaneously with the effective filing date of the present application).
Control systems are currently being employed to control processes for mixing together multiple components in a mixing vessel. An example of such a process is mixing together dry cement and water to form a cement slurry for use in well cementing. Well cementing is a process in which wells that penetrate subterranean formations are formed in the earth, allowing natural resources such as oil or gas to be recovered from those formations. In well cementing, a wellbore is drilled while a drilling fluid is circulated through the wellbore. The circulation of the drilling fluid is then terminated, and a string of pipe, e.g., casing, is run in the wellbore. Next, primary cementing is typically performed whereby a slurry of cement in water is placed in the annulus, which is located between the exterior of the pipe and the walls of the wellbore. Within the annulus, the cement slurry is allowed to set, i.e., harden into a solid mass, to thereby attach the string of pipe to the walls of the wellbore and seal the annulus. Subsequent secondary cementing operations, i.e., any cementing operation after the primary cementing operation, may also be performed. One example of a secondary cementing operation is squeeze cementing whereby a cement slurry is forced under pressure to areas of lost integrity in the annulus to seal off those areas.
Conventional control systems for such a cement mixing process often attempt to control the output flowrate and output density of the mixture exiting the mixing process by controlling the positions of input valves into the system. In the example in which the input valves are an input water valve and an input cement valve, an output slurry density measurement and a total output flowrate measurement are commonly used to control the process. A Proportional-Integral-Derivative (PID) controller may be used to calculate the commanded input water flowrate based on the total commanded input flowrate and the commanded slurry density. It may also be used to calculate the output water flowrate based on the total measured output flowrate and the measured slurry density. Further, a PID controller may be used to calculate the commanded input cement flowrate based on the commanded total input flowrate and the commanded slurry density. Moreover, it may be used to calculate the output cement flowrate based on the total measured output flowrate and the measured slurry density. However, this type of control system has a major drawback in that the response of the water and cement control loops are time lagged. Thus, a change in the water flowrate usually is not observed and corrected for by the cement control loop for some time and vice versa. As a result, oscillations in the density and flowrate may be experienced, especially during transitional phases such as an input disturbance or a commanded change. Another drawback of this control system is that often no densitometer is available to measure the output slurry density, or the output slurry density is ill-conditioned to be used as a control variable (i.e., the value of the density of one component being mixed is very close to the value of the density of the other component being mixed in a two-component system).
The physical system being controlled (e.g. the mixing process) typically exhibits some nonlinear behavior. Using a PI or a PID control system to overcome the physical system nonlinearity results in eigenvalue migration. That is, the eigenvalues, i.e. the parameters that define the control system, are dependent on the operating conditions such as the flowrate and thus experience relatively large shifts in value as the operating conditions change. Unfortunately, the system is a coupled system in that different portions of the system depend upon each other. Thus, fine tuning the control system is typically impossible to accomplish due to the differing time- or frequency-domain responses of the different portions of the system.
In addition to these limitations, the mixing process often experiences disturbances that can lead to inaccuracies in the measurements of the process. Such disturbances include oscillations in the height of the fluid in the mixing vessel, particularly when the mixing vessel is in motion such as in a ship-based mixing process. Another disturbance commonly encountered is that one material, e.g., the dry cement, may become plugged in the pipe being fed to the mixing vessel such that a significant amount of air is required to force the material into the mixing vessel. As such, the fluid in the mixing vessel may contain unaccounted for air.
A need therefore exists for a control system capable of controlling the output flowrate and composition of a mixing process without needing to control or measure the output density of the process. Further, it is desirable to reduce the lag-time of the control system, allowing the process to be monitored and controlled in real time with more accuracy and precision. It is also desirable that the control system be capable of more robustly accounting for disturbances, nonlinearities, and noise that may occur in the mixing process.